A framework to assess the carbon supply–consumption balance in plant roots

Abstract

Plant uptake and transportation of resources, such as water, nutrients and photosynthates, are crucial for plant growth, and a main driver of ecosystem functioning and community responses to environmental changes (Evert, 2006; Ma et al., 2018; Steidinger et al., 2019). Plant roots form a highly branched system with great heterogeneity in structure and function, hierarchically organized from lower order distal roots to higher order basal roots. Roots of the higher branch orders predominantly consist of secondary tissues performing non­absorptive functions, such as transport, storage and anchorage, whereas resource absorption is undertaken only by a few terminal branch orders (usually the first- and second-order roots consisting of primary tissues) (Guo et al., 2008; McCormack et al., 2015). Specifically, in this study, we refer to absorptive roots as fine roots of woody species with intact epidermis and cortex and significant mycorrhizal colonization while lacking secondary growth. Absorptive roots play key roles in plant physiology and ecology, and have received increasing attention in recent studies (Weemstra et al., 2016; Iversen et al., 2017; Valverde-Barrantes et al., 2017; Ma et al., 2018; Kong et al., 2019). Generally, absorptive roots consist of two concentrically arranged and functionally different areas: the stele and tissues outside the stele (ToS, including epidermis, exodermis and cortex). The stele specializes in resource transportation whereas the ToS is responsible for resource uptake and facilitates symbiotic interactions (van Bel & Hafke, 2005; Guo et al., 2008). Mounting evidence shows that the thickness of ToS (i.e. the difference between the root and stele radius) increases much faster than stele radius as absorptive root diameter increases, irrespective of plant phylogeny, mycorrhizal association, growth form or habitat. We, hereafter, refer to this pattern as ‘root structural allometry’ (Fig. 1a) (Valverde-Barrantes et al., 2016; Kong et al., 2019, and references cited therein). Root allometric relationships are important for understanding the functioning and evolution of plant roots (Kong et al., 2017, 2019). This is because anatomical structures of the root stele and ToS are intrinsically linked to a range of key physiological functions, such as respiration (correlated with cortical cell size, Chimungu et al., 2014; Colombi et al., 2019), mycorrhizal associations (correlated with ToS thickness, Kong et al., 2014), absorption and transport of water and nutrients (correlated with ToS thickness and vessel diameter, Steudle & Peterson, 1998) and transportation of photosynthates (correlated with sieve tube diameter, Jensen et al., 2016). Root anatomical structures, such as cortex thickness and vessel diameter, are also phylogenetically conserved (Kong et al., 2014; Valverde-Barrantes et al., 2016). Therefore, root structural allometry could be driven by the physiological processes associated with the stele and ToS. However, little work has been done in elucidating the underlying mechanisms of the root structural allometry in terms of root physiology. Resource transportation in plants occurs through different conduits of the vascular tissues, e.g. vessels in the xylem and sieve tubes in the phloem (van Bel & Hafke, 2005; Jensen et al., 2016). Water and nutrients taken up by the roots are transported upward via vessels, and photosynthates are transported from leaves downward to absorptive roots via sieve tubes to maintain the growth and respiration costs of roots and associated mycorrhizal fungi (Högberg et al., 2001; Mullendore et al., 2010; Kong & Fridley, 2019). Despite the paucity of data on sieve tube size for absorptive roots, one would expect that the diameters of sieve tubes and vessels are linearly correlated because (1) sieve tubes and vessels are ontogenetically related and both are derived from the procambium in the root stele (Evert, 2006); and (2) sieve tubes and vessels are physiologically integrated. For example, humid habitats usually favor plants with larger-diameter vessels, which could transport water and nutrients upwards more efficiently for the production of photosynthates. In turn, these photosynthates are then transported downwards by larger-diameter sieve tubes. As vessel diameter correlates linearly with the diameter of the absorptive roots across plant species (r = 0.79, P < 0.01, Kong et al., 2014), sieve tube diameter may also linearly correlate with root diameter. In our previous work, we used vessel–root diameter relationships as a basis for the ‘nutrient absorption–transportation hypothesis’ (Kong et al., 2017, also see the next section for details). Given the intrinsic connection of sieve tubes with vessels, variation in the diameter of sieve tubes across species could also contribute to the allometric divergence of ToS from stele by larger or more layers of parenchyma in the ToS. Here, we propose a novel framework that uses root structural allometry to explain the balance between carbon (C) supply via sieve tubes in the stele and C consumption by ToS cells. This C balance framework could be integrated with the nutrient absorption–transportation hypothesis (Kong et al., 2017), and further illustrate the inherent links between tissue size, function and C metabolism, which is paramount in understanding plant root evolution, mycorrhizal associations and responses to global climate change. The ToS usually accounts for most of the cross-sectional area in absorptive roots (Guo et al., 2008; Kong et al., 2019). For example, most of the root cross-sectional area (93.2 ± 0.4%, Kong et al., 2014) and total root C content (92.9 ± 1.3%, D. Kong, unpublished) for subtropical trees are covered by the ToS. Therefore, the C consumption rate (including cell construction and maintenance) of the ToS (yc) could approximately represent that of an ‘ideal’ root where both root activity and root diameter are driven by variation in ToS cell number, i.e. yc = kcπrr2, where kc is the C consumption rate coefficient (i.e. C consumption per unit of root cross-sectional area) and rr is the root radius. However, volumetric C flux rate in the sieve tubes (ys) should scale to the fourth power of the sieve tube radius (rs) following the Hagen–Poiseuille law, i.e. ys = ksrs4, where ks is the C flux coefficient of the ideal root. In addition, sieve tube radius as mentioned earlier should scale linearly with absorptive root radius, i.e. rs = krr, where k is the scaling coefficient. Then, volumetric C flux rate can also be expressed as: ys = ksrs4 = ks(krr)4. The intrinsic differences of the scaling relationships of yc and ys with root diameter could lead to an imbalance between C supply by the sieve tubes and C consumption of the ToS. In an earlier study, we described a similar scaling difference between nutrient absorption via mycorrhizal colonization of root ToS (i.e. approximately proportional to the second power of root radius) and nutrient transportation via vessels in the stele (i.e. proportional to the fourth power of root radius) (Kong et al., 2017). This scaling difference calls for a much faster increase of ToS thickness (i.e. faster increase of nutrient absorption via mycorrhiza) over stele radius (i.e. slower increase of nutrient transportation via vessels) to balance nutrient absorption with nutrient transportation (Kong et al., 2017). Likewise, the root structural allometry enables accelerating C consumption via ToS cells and decelerating the C supply by sieve tubes with increasing root diameter, thereby balancing the budget between C sinks and sources in the root system (Fig. 1a). Another important factor influencing C consumption of ToS cells for a single absorptive root is the cell size, which is not accounted for in the earlier estimation of C consumption rates. We find that cell size of root ToS increases linearly with root diameter in woody species (Wang et al., 2019; Supporting Information Fig. S1). Larger cell size in thicker ToS could then reduce the costs of root construction as lower cell numbers, and hence fewer high-cost cell walls would occur compared to roots with equal ToS cell size independent of ToS thickness. Meanwhile, roots with larger ToS cells generally have lower respiration rate because of longer transport time among organelles within the cells as well as higher fraction of cell volume occupied by the low-active vacuole rather than by high-active cytoplasm (Lynch, 2013, 2019; Chimungu et al., 2014; Colombi et al., 2019). Moreover, larger cells could store more nonstructural C or chemical defense compounds for root maintenance and survival (Lux et al., 2004; Sun et al., 2018). This would reduce the resource allocation to cell construction and maintenance, thus reducing the metabolic demands for large ToS cells. In addition, the C costs of mycorrhizal colonization per root cross-sectional area in larger ToS cells may also be lower because of lower activity of the larger host cells and the longer distance of C transport between the host cells and mycorrhizal hyphae. Therefore, the actual C consumption rate of the ToS cells, although higher in thicker than thinner absorptive roots, should scale much slower with root radius than predicted from the earlier quadratic relationship (i.e. yc < kcπrr2). To balance the slower C consumption rate of the ToS cells, one effective way is to decrease the C supply rate from sieve tubes by further decreasing the relative size of sieve tube diameter with respect to root diameter. Phylogenetic evidence demonstrates an evolutionary trend of decreasing diameter of absorptive roots from basal angiosperms to more derived angiosperms since the mid-Cretaceous (Comas et al., 2012; Ma et al., 2018; Valverde-Barrantes et al., 2020). At the same time, using a dataset of woody species (see Table S1), we also found an increasing number of protoxylem poles (i.e. the star-like assembly of vessels in the primary xylem of absorptive roots; see Fig. 1) with increasing root diameter (Figs S2, S3). Here, based on information on currently existing woody species, we propose a decreasing model where absorptive roots evolve (see Model 2 in Fig. 1b2,b4) from thick roots with a higher number of protoxylem poles (e.g. hexarch) to thin roots with a lower number of protoxylem poles (e.g. triarch). Alternatively, we also coin a root evolution model where the number of protoxylem poles is hypothesized to be low in thin absorptive roots and remains constant (or increases; not shown) along the above evolutionary trajectory from thick to thin absorptive roots (see Model 1 in Fig. 1b1,b3). In woody species, phloem tissue in absorptive roots usually occurs between two neighboring protoxylem poles. For absorptive roots of the same diameter, those with a higher number of protoxylem poles would have thinner sieve tubes as less stele area is accounted for by the phloem. Therefore, Model 2 predicts a slower increase of sieve tube diameter, and a slower increase of C supply as root diameter increases (Fig. 1b2 vs b1). This would balance out C supply with C consumption in Model 2 but not in Model 1 (Fig. 1b3 vs b4). Therefore, natural selection for an increase in protoxylem bundles could be a mechanism to balance C supply and consumption between the stele and the cortical tissue in woody species. This in turn also provides new insights into the evolutionary forces shaping protoxylem bundles in plant roots. Our framework for the C supply–consumption balance in plant roots has important implications for our understanding of the relationship between the structure and functioning of absorptive roots. In our model (Fig. 1b), a greater number of protoxylem poles presents a driving force for reducing the C supply through thinner sieve tubes to keep pace with the consumption of C in the root ToS. However, faster increase of ToS thickness will create greater resistance for water uptake into the vascular tissue through the ToS (Steudle & Peterson, 1998). In this case, a higher number of protoxylem poles provides more connection points of protoxylem with the cortex (see Fig. 1b), which facilitates transportation of water and nutrients between the two types of tissue. This could also contribute to nutrient absorption and eventually a better balance with nutrient transportation via vessels (Kong et al., 2017). Therefore, the increased number of protoxylem poles with increasing root diameter in combination with the allometry between root stele and ToS could explain the functional coordination between parenchyma C consumption and xylem transportation (Fig. 2). Furthermore, our C supply–consumption framework provides new insights into the evolution of absorptive roots in terrestrial plants. For instance, in woody angiosperms, there was an evolutionary decrease of root diameter during the Cretaceous period when the atmospheric CO2 concentration substantially declined (Comas et al., 2012). In this perspective, a decrease in ToS thickness can greatly reduce C consumption by reducing C costs in root construction and maintenance (Lynch, 2019) and by shortening root lifespan (Weemstra et al., 2020), which would be positively selected for under declining atmospheric CO2 concentrations. In addition, thinner ToS also decreases hydraulic resistance of water uptake by roots and maximizes surface area in thinner roots. This is consistent with enhanced hydraulic demands of leaf photosynthesis under declining concentrations of atmospheric CO2 (Brodribb & Feild, 2010). Therefore, root structural allometry between the ToS and the stele could have evolved to concomitantly meet the pressures for reducing C consumption and increasing the efficiency of water and nutrient uptake in absorptive roots. These evolutionary insights help to understand how root systems of current species may respond to ongoing climate changes. For example, under elevated levels of atmospheric CO2, plant growth may be limited more severely by nutrients than by C (Luo et al., 2004; Millard et al., 2007). Therefore, under these conditions, plants with thicker absorptive roots may out-compete plants with thinner absorptive roots. This could be due to longer root lifespan (but see Weemstra et al., 2020) and greater amounts of C allocated to and consumed by thicker roots and associated mycorrhizal fungi (Drigo et al., 2010; Dong et al., 2018), via larger-diameter sieve tubes and a higher number of protoxylem poles (Fig. 1b). This then would allow thicker absorptive roots greater nutrient acquisition under elevated levels of atmospheric CO2; an interesting outlook which warrants further testing. While our framework to assess the C supply–consumption balance offers a bridge between root structure and root functioning, it requires more comprehensive sampling and measurements for empirically testing. For example, we need to quantify ToS respiration to determine the C consumption and its relationships with ToS cell size across a wide variety of evolutionary divergent plant species. The C supply rate in root sieve tubes should also be more accurately determined, for example by new methods such as quantum dots (Whiteside et al., 2012) or isotopic labeling (Drigo et al., 2010). Besides interspecific evaluation of our framework, within-species examination, e.g. among different genotypes varying in root ToS and stele size (Chimungu et al., 2014), would also be useful to demonstrate whether C supply is coordinated with C consumption. More attention also needs to be given to anatomical measures of root tissues. In particular, one should focus on the functional rather than the geometric diameter of the sieve tube lumen. Only a portion of the lumen of the sieve tubes, i.e. the callose un-occluded pores in the sieve plates, functions as a conduit for transportation (van Bel & Hafke, 2005; Mullendore et al., 2010). Therefore, C supply rate of the sieve tubes can easily be overestimated when using the geometric diameter of the sieve tubes. In addition, it is also important to consider other sieve structural traits (e.g. the size of sieve plates and diameter of the sieve pores) and factors affecting the sieve tube structure (e.g. protophloem development and fine root turnover rate) which may influence the transport of photosynthates in roots. Moreover, it is important to test root C balance strategies under different climatic and environmental conditions. Our C balance framework assumes that consumed C in roots and associated mycorrhizal fungi comes from recent photosynthates (Högberg et al., 2001; Heinemeyer et al., 2006; Lynch et al., 2013). However, we do not account for the C stored in organs such as stems and coarse roots, which can be important for plant survival in stressful conditions, such as drought or loss of photosynthetic tissues by herbivory (Rocha, 2013). It is worth further testing whether absorptive roots could maintain a C balance with low C supply from stored C and low C consumption by roots and mycorrhizal fungi as both processes are weakened under stressful conditions (Ostonen et al., 2017). Besides nutritional benefits of mycorrhizal fungi, future studies should also take into account the C consumption by different types of mycorrhizas when evaluating the C balance framework. For example, ectomycorrhiza (EM) consist of a thick hyphal mantle and high extramatrical hyphal biomass, and hence give rise to greater mycorrhizal respiration per root length than arbuscular mycorrhiza (AM) (Leake et al., 2004; Heinemeyer et al., 2007; Phillips et al., 2013; Tedersoo & Bahram, 2019). Therefore, we predict that EM roots at a given root diameter would obtain a C balance at higher rates of C consumption by mycorrhizal fungi and higher rates of root exudation (Yin et al., 2014) and, hence, a strong demand for C supply from sieve tubes. Furthermore, to accurately assess the C balance framework, we may need to take into account the C transferred through belowground mycorrhizal networks connecting individual plants (Simard et al., 1997; Klein et al., 2016; Tedersoo et al., 2020). The C transferred via mycorrhizal networks should be added to the C consumption of the donor roots and subtracted from the C consumption of the recipient roots. Finally, we may need to account for C expenditure in root exudates, such as carboxylates, which is particularly important in plants with cluster roots (Lambers et al., 2008). Plants forming cluster roots likely have larger sieve tubes than other plants with similar root diameter because of larger amounts of exudation of carboxylates (Lambers et al., 2008) and the metabolic C costs of producing the exudates (Oilveira et al. 2015). Additionally, plants bearing aerial roots or growing in flooded conditions frequently develop aerenchyma tissue in their roots (Seago et al., 2005), which reduces root C costs as some of the root volume is occupied by air channels (Zhu et al., 2010; Postma & Lynch, 2011; Chimungu et al., 2015; Lynch, 2019). If root structural allometry exists in roots with aerenchyma, the lower C consumption from the ToS cells could be balanced by the lower C supply from the sieve tubes with thinner functional diameter. Moreover, key assumptions of our C balance hypothesis are that both cell size of the ToS and number of protoxylem poles are positively correlated with root diameter. However, these relationships have so far only been observed in woody species. We know little about whether this C balance framework also applies to species with no apparent protoxylem poles, such as grasses. Finally, roots of herbaceous species often have no determinate growth, i.e. root elongation and/or branching are not constrained by the formation of buds or other reproductive structures (Evert, 2006). The excess C consumed by root elongation and/or branching during the formation of reproduction structures should therefore also be taken into account. Taken together, it is necessary to test the C balance hypothesis by designing comparative studies among different plant functional, phylogenetic, and mycorrhizal groups with detailed measurements of root anatomy, which, by doing so, would shed new light on plant root functions and evolution. We thank Dr Hongbo Li, Dr Zhengxia Chen and Dr Hongfeng Wang for their constructive comments on this study. We also greatly thank the handling editor, Dr Colleen Iversen and three anonymous reviewers for their valuable comments and suggestions on earlier versions of this manuscript. This study was supported by the National Natural Science Foundation of China (31670550, 31870522, 31200344 and 41807360), the National Key R&D Program of China (2017YFC1200101), the Scientific Research Foundation of Henan Agricultural University (30500854), Key Platform and Scientific Research Project of Guangdong Provincial Education Department (2019KZDXM028). PK acknowledges support from the Research The no and the and the the of the with from and to the of the The data for Supporting Information and are in Table Fig. between absorptive root diameter and cell size of tissues outside the stele in woody species. Fig. between absorptive root diameter and number of protoxylem poles in the root Fig. Root for the roots of with protoxylem and with Table data of the and are not responsible for the content or of Supporting Information by the than should be to the The is not responsible for the content or of information by the than should be to the for the

Topics & Concepts